Table 5 shows the results from logistic regression of reports of slow healing of any severity. The results indicate a significant treatment effect, mainly among the good compliers. Older age is also a significant predictor in all participants with a trend toward significance among the good compliers, but not among the poor compliers.
Table 6 shows the percentage of participants reporting slow healing at any visit by age, compliance, and drug group. Among older participants, a significantly higher proportion of compliant individuals in the ICS group than in the placebo group reported slow healing (4.3-fold higher proportion; p < 0.005). A similar but nonsignificant trend toward a treatment effect was noted in the compliant younger participants (2.6-fold higher proportion in the ICS group than in the placebo group; p < 0.09). No differences in the proportion of noncompliant participants reporting slow healing were noted between treatment groups in either age category. Source
Correlation Between Reported Bruising and Impairment in Adrenal Function and Reduction in BMD
In an ancillary study to LHS II involving a subset of 221 participants in 5 of the 10 centers, the HPA axis was evaluated at baseline, year 1, and year 3. The development of significantly impaired adrenal function was rare (1.8%), and no consistent or meaningful differences were noted between the presence or absence of bruising vs change in basal or stimulated morning serum cortisol level, irrespective of whether analyses were performed in all participants or were restricted to only the good compliers.
In another LHS II ancillary study, BMD of the femoral neck and the lumbar spine was measured by dual-energy radiograph absorptiometry in a subset of 412 participants (ICS group, 201 participants; placebo group, 211 participants) at baseline and after 1 year and 3 years of treatment. In addition, the serum osteocalcin level was measured at baseline, 3 months, 1 year, and 3 years. After 3 years of treatment, BMD was lower in the ICS group than in the placebo group at both the hip and lumbar spine. Osteocalcin levels were also reduced to a greater extent in the ICS group than in the placebo group at 3 months (p = 0.001) and 1 year (p < 0.04) [PD Scanlon, MD; personal communication; September 3, 1999). Thirty-one participants in the ICS group (18%) and 13 participants in the placebo group (7%) had a a 6% decrease in femoral neck BMD compared with baseline, while 14 participants in the ICS group (9%) and 9 participants in the placebo group (6%) had a a 6% decrease in lumbar spine BMD between baseline and year 3. Among participants who experienced a 6% declines in BMD of either the femoral neck or the lumbar spine, the proportion reporting bruising was not significantly different in the ICS group vs the placebo group, irrespective of inhaler compliance. Furthermore, no relationship was found between reported bruising and changes in BMD, either among all participants or in good compliers only.
Table 5—Predictors of Slow Healing
|Predictor||All Participants (n = 1,086)||a 6 Puffs/d (n = 620)||Up to 6 Puffs/d (n = 466)|
|P Estimated (SE)||p Value||P Estimated (SE)||p Value||P Estimated (SE)||p Value|
|Age (10-yr increments)||0.539 (0.232)||0.0202||0.495 (0.288)||0.0850||0.558 (0.406)||0.1699|
|Drug (ICS vs placebo)||0.370 (0.157)||0.0182||0.654 (0.208)||0.0017||-0.129 (0.264)||0.6247|
Table 6 —Percentage of Participants Reporting Slow Healing at Any Visit By Age, Compliance, and Drug Group
|Dosage||Age < 56 Years||Age a 56 Years|
|a 6 puffs/d||7.4(3.0-11.8)||2.8 (0.1-5.5)||9.3 (4.8-13.8)||2.2 (0.1-4.4)|
|< 6 puffs/d||1.5 (0-3.6)||1.8 (0-4.2)||4.3 (0.6-8.1)||5.7(1.3-10.2)|